Search in the Abstract Database

Search Abstracts 2011

* = Presenting author

P082. Diagnostic accuracy of confocal fluorescence microscopy in diagnosing dysplasia in patients affected by long-standing ulcerative colitis

A. Rispo1, F. Castiglione1, F. Maione2, S. Siciliano3, D. Esposito2, A. Testa1, F. Salvatori3, F. Sasso1, C. D'Onofrio1, G.D. De Palma2

1Gastroenterology - University “Federico II” of Naples, Naples, Italy; 2Surgery and Advanced Technologies - University “Federico II” of Naples, Naples, Italy; 3Surgery and Advanced Technologies - University “Federico II” of Naples, Naples, Italy

Background: Patients affected by long-standing ulcerative colitis (UC) are candidate to surveillance colonoscopy/histology in view of the increased risk of colon cancer/dysplasia. Previous studies have shown that detection of dysplasia is significantly increased by chromoendoscopy, while data about the use of confocal fluorescence microscopy (CFM) are still lacking.

Aim: To evaluate the diagnostic accuracy of CFM in detection of dysplasia in patients affected by long-standing UC.

Methods: From March 2009 to October 2010 we prospectively performed surveillance colonscopy in 39 patients (20 male, 15 female; median age 52 years) affected by long-standing UC (19 extensive colitis, 16 distal colitis). Also, in presence of macroscopic areas suspected for dysplasia (on flat mucosa or mass) both focal contrasted indaco carmine endoscopic assessment and CFM (Cellvizio®) were performed. Colic mucosal biopsies and histology, utilised as gold standard, were assessed randomly and on visible lesions, in accordance with the ECCO guidelines. Diagnostic accuracy of CFM in prediction of dysplasia when compared to standard histology was assessed by StatsDirect statistical software.

Results: 14 out of the 39 patients (35%) showed mucosal macroscopic alterations suspected for presence of dysplasia, needing chromoendoscopic and CFM evaluation. In 5 macroscopically suspected cases (35%) the presence of dysplasia was confirmed by both histology (3 flat dysplasia; 2 DALMs) and the following surgery. No dysplasia/cancer was found on all outstanding random biopsies. The diagnostic accuracy for detection of dysplasia of CFM respect to the standard histology was: sensitivity 100%, specificity 92%, positive predictive value 83%; negative predictive value 100%.

Conclusion: Confocal fluorescence microscopy is an accurate tool for the detection of dysplasia in long-standing UC with optimal values of specificity and negative predictivity. The scheduled combined application of chromoendoscopy and CFM could maximize the endoscopic diagnostic accuracy for diagnosis of dysplasia in UC patients, so limiting the need for biopsies. Further studies including a wider population are needed to confirm our suggestion.